Corona Sweden : Probability of dying per age group

Just for fun, hacked a simple Bayesian Model* to figure out how the probability of dying has changed for a few age groups Jan-Jun 2020, compared to the average of the same period 2015-2019.  (In case anyone wonders why on earth I use such weird age group as 0-64, the simple explanation is that that’s how the data, including the prel. numbers for 2020,  are presented by

So, how has the probability changed, according to the model…?  Not that much, it turns out: for the 0-64 age group, mean has moved from 0.00048 to 0.00055; for 65-79: 0.0075 to 0.0083, for 80-89: 0.0325 to 0.0365 and for 90+ from 0.103 to 0.114

The model’s beliefs are very much in line with the analytically calculated numbers, only difference is for the 0-64 age group, where according to the model is a slightly increased probability of dying 2020, whereas the actual data shows that the numbers for that age group are almost identical, in fact with a tiny advantage for 2020.



*the gist of the Bayesian model:

# model:
# dead ~ Binomial(population,p)
# p = logit(alpha[age_idx] + beta * year)
# alpha[age_idx] ~ Normal(0,10)
# beta ~ Normal(0,10)

About swdevperestroika

High tech industry veteran, avid hacker reluctantly transformed to mgmt consultant.
This entry was posted in Bayes, Data Analytics, Epidemics, MCMC, PYMC and tagged , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s